Scenarios for cellular agriculture

By JesseClifton @ 2016-06-20T21:40 (+15)

Cellular agriculture - the production of animal products in cell cultures - has the potential to massively reduce suffering by replacing systems of animal exploitation. In this post I explore different scenarios for the development of cellular agriculture, and the implications for the effort to end factory farming and commercial fishing. (Hereafter I write “factory farming” for brevity, without forgetting the countless billions of wild sea creatures killed for food each year.)

Much of what I say also applies to realistic plant-based substitutes, like those produced by Beyond Meat. The difference is that these seem easier to achieve (Beyond Meat already has mass-market products that are convincing to at least some people), but less likely to replace animal products. The conclusions are the same.

Throughout the article I will refer to the Open Philanthropy Project's (Open Phil) December 2015 write-up on animal product alternatives, which I strongly recommend reading in its entirety.

Summary

Despite optimism in some circles over cellular agriculture, the path to cost-competitive cultured animal products (CAPs) is hazy and the timeline disputed. In any case, cellular agriculture alone cannot be relied on to end factory farming; there are serious psychological and political hurdles in even the best-case scenarios for CAP technology. More difficult still are the not-improbable scenarios in which cost-competitive CAPs arrive only on some markets, or only for certain kinds of animal product. These may leave factory farming largely intact even with complete substitution of CAPs for factory farmed products. The conclusion is that animal activism remains essential to reducing animal suffering, and that the promise of cellular agriculture should not count against factory farming in cause prioritization.

Forecasts

An Atlantic article - published August 6, 2013, the day after the debut of the first lab-grown burger13 - featured this chart recording predictions about in vitro meat made by scientists and journalists. (The beginning of each horizontal bar is the prediction date and the end is the date by which the event is predicted to occur. Some predictions used the vague language “in a few years” - these were coded as “three years from now”.)14

Several predictions about the timeline of cellular agriculture have been made since 2013. In 2014 it was reported that the in vitro dairy company Muufri predicted “most of us will be drinking artificial milk in 100 years”.15 In 2015 conversations with Open Phil, leading in vitro meat researcher Mark Post predicted that cost-competitive cultured meat would arrive within 7-10 years, while “a scientist with 18 years experience in the tissue engineering industry” said that "Without a major technological breakthrough, it seems very unlikely that cost-competitive cultured meat will be available in the next 10-15 years.”12

Open Phil also reports three predictions about the cost of in vitro meat (reproduced from [12]):

Estimate by:	Cost of cultured meat (USD)	Assumed manufacturing volume	Year
Vandenburgh	$5M / kg	Small-scale production in laboratories	2004
Exmoor	€3300 – 3500 / tonne (€3.3 – 3.5 / kg)	Scaled-up to large volume	2008
Van der Weele and Tramper	€391 / kg assuming typical media cost of €50,000 / m³. One estimate of the lowest possible cost of media is €1,000 per m³, but we do not know what this estimate is based on. Plugging this assumption into Van der Weele's model would imply that €8 of media is needed for 1 kg of meat	Scaled-up to large volume	2014

(In a 2015 interview, Mark Post estimated the price of producing an in vitro hamburger at less than $1218 - down from $325,000 in 201313 - but it isn’t clear what this figure is based on.)

In considering the likelihood of cost-competitive cellular agriculture, Open Phil discusses two (relatively) similar cases: tissue engineering company Organogenesis, which used a process with certain similarities to meat culture to create wound care skin grafts, and synthetic biofuel company Amyris. On the basis of these companies’ failures to achieve mass-market costs, along with a “holistic assessment of the challenges involved in reducing the cost of cultured meat, [and] discussion with scientists who have experience with cell cultures and tissue engineering”, Open Phil concludes that they regard “developing cost-competitive cultured meat products as extremely challenging, and we have been unable to find any concrete paths forward that seem likely to achieve that goal.”12

I am not in a position to offer my own forecasts, and in such uncertain domains I favor strategies which are robust to a variety of outcomes over those which try to anticipate what will happen. (See this post on the advantages of scenario planning over forecasts in AI progress, another uncertain domain unfolding over similar time horizons.) Nevertheless, I think these divergent opinions serve as evidence against the inevitability of mass-market cellular agriculture, especially in the next few decades.

Scenarios for 2050

In this section I consider several scenarios for the state of cellular agriculture worldwide in 2050. The year is chosen somewhat arbitrarily, but seems reasonable since 1) it is in the not-too-distant future and 2) predictions in the previous section have cost-competitive CAPs arriving by 2035, which leaves 15 years for the realization of the various possibilities.

Cost-competitive versions of all animal products available worldwide
Cost-competitive cultured versions available for some products
No cost-competitive cellular agriculture

All animal products have cost-competitive, cruelty-free in vitro versions on the market

In the best-case scenario, every type of (sentient) animal product now consumed - every kind of flesh, dairy, eggs, collagen, leather, fur, silk, honey, etc. - has a cost-competitive exact substitute available worldwide, and these substitutes require no harm to animals*. (In vitro meat currently requires fetal bovine serum, which is extracted from the heart of an unborn calf1. Even a scenario in which cruelty-free methods were not available would be a substantial improvement, given that vastly fewer animals would be necessary to meet animal product demand2, but may still involve some animals kept in factory farm-like conditions subjected to painful procedures.) Not only does this directly cause reductions in suffering as people switch from in vivo to in vitro animal products, it places the animal rights movement in a much stronger position to argue for legal bans on animal exploitation, as it becomes less tenable to regard animal exploitation as “necessary” (I use scare quotes because animal exploitation is, of course, already unnecessary in the great majority of cases).

Pushing for legal bans will be crucial, since we cannot count on demand alone to dispense with factory farming. First, some animal products involving horrific cruelty already have convincing substitutes - faux fur, for instance - but remain on the market. Second, growing concern over genetically modified organisms and a trend toward “natural” eating suggests that a significant proportion of consumers are likely to be opposed to artificial meat on the grounds of its unnaturalness. Lastly, the few opinion polls conducted on the subject reveal a widespread aversion to artificial meat**:

2014 Pew Survey: 78% of Americans say they would not “eat meat grown in a lab”3
2012 YouGov survey: 62% of British say they “probably would not eat” artificial meat4
2005 EU survey: 54% of Europeans would “never” approve of artificial meat5

Now, it is entirely possible that disapproval of cultured meat will abate. In vitro fertilization is one example of an accepted practice which was once opposed by a significant fraction of the population due to its “unnaturalness”, though according to a Gallup poll taken in 1978—the year the first “test-tube baby” was born—only 28% opposed IVF, much lower than the numbers opposed to in vitro meat21. But the transition from distaste to acceptance is by no means assured: for instance, GMO opposition seems to be gaining in popularity, with growing demand for non-GMO foods22 and major food companies getting rid of GMO ingredients23 and labeling GMO-containing products24.

I get the impression that many EAs believe the total replacement of factory farming by cellular agriculture is inevitable. But humans have long inflicted extreme suffering on animals for utterly trivial reasons. I worry that abnormally rational and altruistic people may underestimate the persistence of animal product demand even when ethical substitutes require the tiniest psychological burden (vague discomfort, change of routine, etc.). Environmental protection, biosecurity, and health—all of which may be improved by cellular agriculture—appear to be similarly insufficient motivators.

Assuming worldwide availability, it is unclear which countries will be most and least resistant to CAPs. Western countries (especially Europe), along with a few non-Western countries like Israel, have the highest levels of concern for animal welfare and rights. But, as we have seen, Westerners have already expressed considerable reluctance to eat cultured meat. In China, whose animal product consumption and production is massive and growing6,7,8, there is both little concern for animals9,10 and extensive opposition to GMOs11. This is quite disturbing if attitudes toward GMOs predict acceptance of cellular agriculture.

What would an animal rights movement look like under this scenario? Given that we can’t expect the market to drive out animal exploitation on its own, the movement should be emboldened rather than contented. More people are likely to sympathize with animal rights, as the possibility of using CAPs can eliminate the cognitive dissonance between concern for animals and participation in their torture. With cost-competitive animal product replacements available to the public, governments are more likely to consider major reforms like animal personhood and bans on entire industries.

But as with economics, there is no guarantee that politics will bring an end to factory farming. First, even a freshly empowered animal rights movement may be unable to make the end of animal exploitation a salient political issue. Not only do many people believe that animal suffering should not be bothered with until human problems are solved, many are simply indifferent to the suffering of animals and would be unconcerned with ending animal exploitation even if cellular agriculture rendered it “unnecessary”. Second, bans on animal exploitation will still require fighting the extremely powerful agricultural industry (assuming the large part of the animal agriculture industry does not switch to cellular agriculture). If cellular agriculture poses a serious threat to conventional agriculture, we can expect the wealthy animal product industry to pour resources into media campaigns attempting to discredit “fake”, “unnatural” CAPs and into lobbying efforts to prevent the reforms sought by animal rights activists.

In summary, the abolition of factory farming is rife with challenges in even the most optimistic scenario for cellular agriculture. The animal rights movement will have considerably more leverage, but will still face the formidable obstacles of distaste, apathy, and the behemoth of the animal exploitation lobby. Winning this fight in a single country will be difficult enough. Winning in dozens of countries is far from inevitable, cellular agriculture or not.

Cost-competitive cellular agriculture for only certain products

In this scenario, we achieve mass-market cultured replacements for only certain types of animal products. According to Open Phil’s article on animal product replacements, cultured egg whites may be significantly easier to produce than cultured ground meat, which in turn may be easier than slab meat—large pieces of muscle tissue, including chicken breasts and fish filets. Moreover, chicken and fish, which constitute the vast majority of animals killed for food, appear to be a neglected area of in vitro meat research relative to their numbers. Israel’s Modern Agriculture Foundation and a new project at NC State27 are the only groups I have been able to find working on cultured chicken, and besides Mark Post’s mention of a “team in Queensland, Australia” in his conversation with Open Phil I have seen no active projects on cultured fish12. It is not implausible, then, that we will see indefinite lags between the arrival of different CAPs to the mass market.

Consider a scenario in which researchers have developed a number of in vitro animal products but are unable to develop slab chicken meat. In this case, even the complete replacement of farm animal products with available cultured products leaves tens of billions of chickens subjected to extreme suffering each year.

Cultured replacements for different types of animal food products would lead to some reduction in chicken suffering, as a portion of consumers would replace some or all of their chicken flesh with cultured meat. If cultured meat is cheaper than farmed animal flesh, substitution may drive significant reductions in the number of chickens farmed for food. But today there is still high demand for expensive meats even when comparatively cheap ones are available (for example, according to 2010 USDA estimates, the cross-price elasticity of chicken with respect to beef was 0.02; wrt pork, 0.31; and wrt fish, 0.23, indicating only weak to moderate substitution between meats17). So even in the optimistic case that cultured meats become cheaper than their factory-farmed counterparts, we can expect factory farming to persist.

If cultured meat is at least as expensive as farmed animal meat, it is doubtful that a substantial fraction of consumers would substitute, given the current unwillingness to replace animal products with similar vegan substitutes. (Granted, for the typical consumer the similarity between chicken and pig is probably greater than that between e.g. cow’s milk and soy milk, so there is reason to think substitution between meats would be somewhat more common.) With the availability of some cultured meats, animal rights activists would be in a better position to argue that farming chickens is “unnecessary”, but again this seems unlikely to sway a critical mass of the population given that many would be unwilling to accept substitutes for chicken flesh.

No cost-competitive cellular agriculture

In this scenario, cost-competitive CAPs are never achieved. Such products may appear in specialty stores or restaurants, but remain too expensive to replace conventional animal products in large numbers. Animal rights activists must continue to fight the extremely difficult battle of abolishing massive animal exploitation while no exact substitutes for animal products exist.

Implications

Animal Activism

To end massive animal suffering at the hands of humans, we will need a robust animal rights movement under any scenario for cellular agriculture technology, even the most optimistic. We cannot expect demand alone to drive the abandonment of factory farming.

This means that animal activists who are sanguine about the prospects for cellular agriculture must not get complacent. Even if we are highly confident about the worldwide availability of cost-competitive cellular agriculture, we must build as much momentum as possible now to be able to capitalize on whatever political and psychological leverage is generated by these developments. Mass-market CAPs will only be the beginning of a serious worldwide animal liberation movement.

What do the different possibilities imply for welfarism versus abolition? In another post I argued that given the great uncertainty as to which methods are most effective in a particular time or place, animal activists should cultivate a diverse portfolio of approaches to reducing animal suffering. This uncertainty is only compounded by uncertainty over the future of cellular agriculture—and which methods will be most effective under the various scenarios—so I believe that diversity remains the best structure for animal activism. (This does not mean the mix of strategies which constitutes animal activism at present is optimal. For instance, the non-negligible possibility of cost-competitive cellular agriculture may imply the need for a greater balance of liberationist messaging, as this kind of approach will be strengthened if CAPs are available to replace factory farming.)

Funding

As approaches to reducing animal suffering, cellular agriculture and animal activism compete to some extent for funding. There is some reason to believe overlap in potential funders is not too great, since major investors in cellular agriculture will include venture capitalists (like New Crop Capital, which will invest $25 million over 5 years in animal product replacement firms including CAP companies Memphis Meats and Gelzen19) and government agencies (like the Dutch SenterNovem which funded cultured meat research from 2005-0920) that wouldn’t otherwise invest in animal activism.

Nevertheless, many altruists will find themselves deciding between donations to cellular agriculture research and to animal activist groups. Animal activism will be vital to reducing animal suffering whatever the availability of CAPs. This means that the relative value of donations to cellular agriculture research and animal activism at any given point will largely be constrained by how promising cellular agriculture appears at the time, and its need for funding.

Cause Prioritization

EAs who do not prioritize factory farming on the grounds that it will inevitably be solved by cellular agriculture should reconsider. Even if cost-competitive CAPs are as likely as they believe (and they may be overconfident; see the Forecasting section), they are probably insufficient to end factory farming. Animal rights movements throughout the world will need talent and funding to translate their new strength into major reductions in suffering. (In fact, the value of working to end factory farming arguably increases with one’s confidence in mass-market cellular agriculture, since cellular agriculture will not end factory farming alone but can make campaigning to end factory farming far more tractable.)

My views on this topic have been shaped my research on animal welfare/rights in different countries and throughout time (e.g. Animal welfare and rights in India, Timeline of animal welfare and rights), on the history of cellular agriculture, and on social movements other than animal activism (fairly brief).

Thank you to Buck Shlegeris, Carl Shulman, Claire Zabel, Jacy Reese, Michael Dello-lacovo, Michael Dickens, and Vipul Naik for their feedback on this post. Thanks to Vipul Naik for funding my work on this post (the views expressed are mine).

Notes

*Some argue that the best case scenario for animal farming is instead that in which we still have farm animals, but they lead lives worth living. It is not clear where cellular agriculture fits in this picture. In any case, I hope you will agree that cost-competitive, cruelty-free cellular agriculture is at least a locally optimal outcome.
**Less rigorous polls have found more support. In a survey conducted by The Vegan Option blog, 28% of omnivores answered “No” to “Would you eat lab meat?” and 36% answered “Maybe/I’m not sure”25. In a poll on Sam Harris’ Twitter, only 17% said they would not switch to cultured meat26.

Sources

undefined @ 2016-07-25T06:22 (+4)

Given the large number of animals who would be spared the suffering involved in factory farming by making the transition to CAP happen slightly more quickly, even a small chance of the additional funding making the difference would justify donating to CAP research over vegan advocacy.

For example, let's consider someone with $1,000,000 to give away. According to the ACE impact calculator, if you spent that amount of money purchasing leaflets for volunteers to distribute, you would spare 4,000,000 animals from factory farming. What about if you spent it on grants for academics researching cultured chicken meat? If you think that there is at least a one in one hundred chance that you will make each point in the transition away from eating factory farmed chicken in the United States occur just one month earlier, then you would expect to spare at least 7,000,000 chickens from factory farming (8,800,000,000 chickens slaughtered in the US last year divided by 12 months in a year multiplied by 0.01 probability). I did leave out wild caught fish in my estimate of the total number of animals spared by leafleting (which makes it an underestimate), but I also left out factory farmed chickens outside the United States in my estimate of the total number of animals spared by CAP research (which also makes it an underestimate).

Because of the large number of animals that would be saved by even a slightly quicker transition away from factory farming, I think the top donors focused on animals should seriously consider grants for CAP research when the research is likely to meaningfully advance the transition to CAP.

undefined @ 2016-07-28T04:55 (+1)

One potential counterargument to this is that the primary constraint on the adoption of cultured animal products is a lack of demand rather than a lack of supply. If nobody wants cultured animal products for the next half century, then bringing them to market at a competitive price in one decade instead of two decades has no effect for the next half century. There does seem to be some evidence for this view: http://www.pewinternet.org/files/2014/04/PI_2014.04.16_TechFuture_driverless_cars.png

However, while more animal advocacy now could increase the demand for CAP thirty years from now, introducing CAP a decade from now instead of two decades from now will also do that. Specifically, technology adoption tends to follow a pattern with some people adopting it immediately, some people adopting it after a few years, some people adopting it after most of their friends have adopted it, and some people never adopting it. Because earlier introduction of CAP will shift the adoption date for more than just those who adopt it immediately, it may have a very significant effect on demand several decades from now.

undefined @ 2016-07-04T15:07 (+3)

Great post!

I think it is good to keep in mind that even though the raw materials for plant-based meat substitutes are significantly cheaper than the animal products, still these meat substitutes are much more expensive per calorie even though we have significant economies of scale now.

I agree that cost competitiveness is not enough. But if meat substitutes actually tasted better or were more healthful, this could be a scenario with strong spontaneous adoption.

undefined @ 2016-06-20T23:19 (+2)

This is a very nice post. I very much agree with these statements:

"This means that the relative value of donations to cellular agriculture research and animal activism at any given point will largely be constrained by how promising cellular agriculture appears at the time, and its need for funding."
"This does not mean the mix of strategies which constitutes animal activism at present is optimal. For instance, the non-negligible possibility of cost-competitive cellular agriculture may imply the need for a greater balance of liberationist messaging, as this kind of approach will be strengthened if CAPs are available to replace factory farming."

I'm not sure I entirely agree about the second sentence here, if you're talking about the absolute number of funders: "As approaches to reducing animal suffering, cellular agriculture and animal activism compete to some extent for funding. There is some reason to believe overlap in potential funders is not too great." Though there's a lot of VC money that appears ready to fund private ventures, I believe a large amount of the funding going to research (e.g., what New Harvest recommends or to New Harvest) is coming from those individuals who are choosing between organizations and efforts to reduce animal suffering. My sense though is that the Dutch government's prior funding outweighs all of that in sum though.

Small correction: New Crop Capital has not invested $25 million yet. They've invested $5 million of $25 million and will invest the remaining $20 million over the next four years.

undefined @ 2016-06-20T23:56 (+1)

Thanks a lot! I've made the correction you pointed out.